""" Extracting Shock Data for Feature Extraction - Initial Plots for BPF ==================================================================== Fetching shock instances (including symptoms that indicate severe dengue) and plotting the data. Processing of data will follow this step """ #%% ## Importing Libraries, Definitions and Data Loading # Generic import os import numpy as np import pandas as pd import matplotlib.pyplot as plt from datetime import datetime import glob import plotly.graph_objects as go from plotly.subplots import make_subplots import plotly.io as pio Terminal = True SQI_clinical_file = r'..\..\..\..\OUCRU\Outputs\Complete_SQIs_with_Clinical.csv' Raw_signals = r'..\..\..\..\OUCRU\Outputs\Raw_signals.csv' #Loading SQI Matched with clinitcal to the Dataframe SQI_C = pd.read_csv(SQI_clinical_file) Raw = pd.read_csv(Raw_signals) if Terminal: print("\n SQI with Clinical Match:") print(SQI_C) print("\n Raw Signals") print(Raw) #List of events explored event = ['event_shock', 'reshock24','diagnosis_admission',\ 'ascites', 'respiratory_distress', 'ventilation_cannula', \ 'ventilation_mechanical', 'ventilation_ncpap', 'bleeding_severe', \ 'cns_abnormal', 'liver_mild', 'pleural_effusion', 'skidney'] shock_ad = 'shock_admission' #excluding it from the prior list as we are treating this differently #Study_no list for ease of use patient_list = ['003-2162'] #['003-2009', '003-2012','003-2023','003-2028','003-2103','003-2104','003-2109', '003-2110', '003-2162'] #%% # Turning signal window rows with an event to "keep = True" SQI_C['keep'] = False for i in range(len(event)): event_s = event[i] SQI_C['keep'][SQI_C[event_s] == True] = True print("\n Total ", event[i], " events:") #Optional Save #SQI_C.to_csv(r'..\..\..\..\OUCRU\Outputs\Complete_SQIs_with_Clinical_keep.csv') # %% #Plotting IR_ADC BPF and PLETH BPF on interactive graphs, together with the events (i.e. event of shock and whether or not the patient was admitted with shock) for i in range(len(patient_list)): #Plotting IR_ADC BPF signals with events #fig = make_subplots(rows=2, cols=1) title_str = 'IR_ADC_BPF over Time for patient ' + str(patient_list[i]) fig = go.Figure(layout_title_text = title_str) fig.add_trace(go.Scatter(x = Raw.PPG_Datetime[Raw.study_no == patient_list[i]], y = Raw.IR_ADC_bpf[Raw.study_no == patient_list[i]], name='IR_ADC'))#,),row=1,col=1) for index, row in SQI_C[(SQI_C.study_no == patient_list[i]) & (SQI_C.keep == True)].iterrows(): fig.add_vline(x = row.PPG_w_s, line_width=3, line_dash="dot", line_color="red")# annotation_text="Shock")#, annotation_position="top left", annotation_font_size=20, annotation_font_color="red") if not row.empty: fig.add_trace( go.Scatter(mode='markers', x=[row.PPG_w_s], y=[Raw['PLETH'][(Raw.study_no == patient_list[i]) & (Raw.PPG_Datetime >= row.PPG_w_s) & (Raw.PPG_Datetime < row.PPG_w_f)]], marker=dict(color='red', opacity=1), name = "Shock" )) if not SQI_C[(SQI_C.study_no == patient_list[i]) & (SQI_C.shock_admission == True)].empty: fig.add_annotation( xref="x domain", yref="y domain", # The arrow head will be 25% along the x axis, starting from the left x=0.01, # The arrow head will be 40% along the y axis, starting from the bottom y=0.01, text="Admitted with Shock", showarrow=False, font=dict( family="Courier New, monospace", size=20, color="RED" ) ) fig.update_xaxes(rangeslider_visible=True) fig.update_xaxes(title_text="Date") fig.update_yaxes(title_text="IR_ADC") fig.update_layout(showlegend=True) # #IGNORE, this is to save locally # imagepath1 = "D:\FILES\Desktop\Dissertation ICL\OUCRU\Outputs\Images_BPF\PNG" # imagepath2 = "D:\FILES\Desktop\Dissertation ICL\OUCRU\Outputs\Images_BPF\SVG" # imagepath3 = "D:\FILES\Desktop\Dissertation ICL\OUCRU\Outputs\Images_BPF\HTML" # img_title1 = os.path.join(imagepath1,title_str) # img_title2 = os.path.join(imagepath2,title_str) # img_title3 = os.path.join(imagepath3,title_str) # img_s1 = img_title1 + ".png" # img_s2 = img_title2 + ".svg" # img_s3 = img_title3 + ".html" # fig.write_image(img_s1) # fig.write_image(img_s2) # fig.write_html(img_s3) fig.show() #Plotting PLETH BPF and events #fig = make_subplots(rows=2, cols=1) title_str2 = 'PLETH_BPF over Time for patient ' + str(patient_list[i]) fig2 = go.Figure(layout_title_text = title_str2) fig2.add_trace(go.Scatter(x = Raw.PPG_Datetime[Raw.study_no == patient_list[i]], y = Raw.PLETH_bpf[Raw.study_no == patient_list[i]], name='PLETH'))#,),row=1,col=1) for index, row in SQI_C[(SQI_C.study_no == patient_list[i]) & (SQI_C.keep == True)].iterrows(): fig2.add_vline(x = row.PPG_w_s, line_width=3, line_dash="dot", line_color="red")#, annotation_text="Shock")#, annotation_position="top left", annotation_font_size=20, annotation_font_color="red") if not row.empty: fig.add_trace( go.Scatter(mode='markers', x=[row.PPG_w_s], y=[Raw['PLETH'][(Raw.study_no == patient_list[i]) & (Raw.PPG_Datetime >= row.PPG_w_s) & (Raw.PPG_Datetime < row.PPG_w_f)]], marker=dict(color='red', opacity=1), name = "Shock" )) if not SQI_C[(SQI_C.study_no == patient_list[i]) & (SQI_C.shock_admission == True)].empty: fig2.add_annotation( xref="x domain", yref="y domain", # The arrow head will be 25% along the x axis, starting from the left x=0.01, # The arrow head will be 40% along the y axis, starting from the bottom y=0.01, text="Admitted with Shock", showarrow=False, font=dict( family="Courier New, monospace", size=20, color="RED" ) ) fig2.update_xaxes(rangeslider_visible=True) fig2.update_xaxes(title_text="Date") fig2.update_yaxes(title_text="PLETH") fig2.update_layout(showlegend=True) # #IGNORE, this is to save locally # img_title4 = os.path.join(imagepath1,title_str2) # img_title5 = os.path.join(imagepath2,title_str2) # img_title6 = os.path.join(imagepath3,title_str2) # img_s4 = img_title4 + ".png" # img_s5 = img_title5 + ".svg" # img_s6 = img_title6 + ".html" # fig2.write_image(img_s4) # fig2.write_image(img_s5) # fig2.write_html(img_s6) fig2.show() #MULTIPLE plots same figure #fig.append_trace(go.Scatter(x = Raw.PPG_Datetime[Raw.study_no == '003-2009'], y = Raw.PLETH[Raw.study_no == '003-2009'], name='Pleth'), row=2, col=1) #for index, row in SQI_C[(SQI_C.study_no == '003-2009') & (SQI_C.keep == True)].iterrows(): # fig.add_vline(x = row.PPG_w_s, line_width=3, line_dash="dash", line_color="red", row=2, col=1) #fig.update_yaxes(title_text="Infrared", row=1, col=1) #fig.update_yaxes(title_text="Pleth", row=2, col=1) # fig.update_xaxes(rangeslider_visible=True, row=1, col=1) # fig.update_xaxes(title_text="Date", row=1, col=1) # fig.update_xaxes(rangeslider_visible=True, row=2, col=1) # fig.update_xaxes(title_text="Date", row=2, col=1) #fig.update_layout(height=1000, width=1000, title_text=" IR_ADC and PLETH Plots for Patient 003-2009") # %%